A comparative study of state-of-the-art deep learning architectures for rice grain classification
Accurate and efficient automated rice grain classification systems are vital for rice producers, distributors, and traders, offering improved quality control, cost optimization, and supply chain management. They also hold the potential to aid in the development of rice varieties that are more resist...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-03-01
|
| Series: | Journal of Agriculture and Food Research |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666154323003976 |
| _version_ | 1827324731853897728 |
|---|---|
| author | Farshad Farahnakian Javad Sheikh Fahimeh Farahnakian Jukka Heikkonen |
| author_facet | Farshad Farahnakian Javad Sheikh Fahimeh Farahnakian Jukka Heikkonen |
| author_sort | Farshad Farahnakian |
| collection | DOAJ |
| description | Accurate and efficient automated rice grain classification systems are vital for rice producers, distributors, and traders, offering improved quality control, cost optimization, and supply chain management. They also hold the potential to aid in the development of rice varieties that are more resistant to disease, pests, and environmental stress. While most existing studies in the rice classification domain rely on traditional machine-learning techniques that necessitate feature extraction engineering processes, our research explores the effectiveness of novel deep-learning models for this task. We evaluated the performance of various contemporary deep-learning models, including Residual Network (ResNet), Visual Geometry Group (VGG) network, EfficientNet, and MobileNet. These models were tested on a dataset comprising 75,000 images, classified into five different rice categories. We assessed each model using established evaluation metrics such as accuracy, F1 score, precision, recall, and per-class accuracy. Our findings showed that the EfficientNet-based model delivered the highest accuracy (99.67%), while the MobileNet-based model excelled in the speed of classification (2556 s). We concluded that, compared to traditional machine learning methods, the models employed in our study are highly scalable and capable of managing large volumes of complex data with millions of features and samples. |
| first_indexed | 2024-03-07T14:00:16Z |
| format | Article |
| id | doaj.art-9ee0be1144094200b5199073f0453811 |
| institution | Directory Open Access Journal |
| issn | 2666-1543 |
| language | English |
| last_indexed | 2024-03-07T14:00:16Z |
| publishDate | 2024-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of Agriculture and Food Research |
| spelling | doaj.art-9ee0be1144094200b5199073f04538112024-03-07T05:29:27ZengElsevierJournal of Agriculture and Food Research2666-15432024-03-0115100890A comparative study of state-of-the-art deep learning architectures for rice grain classificationFarshad Farahnakian0Javad Sheikh1Fahimeh Farahnakian2Jukka Heikkonen3Corresponding author.; Department of Computing, University of Turku, 20500, Turku, FinlandDepartment of Computing, University of Turku, 20500, Turku, FinlandDepartment of Computing, University of Turku, 20500, Turku, FinlandDepartment of Computing, University of Turku, 20500, Turku, FinlandAccurate and efficient automated rice grain classification systems are vital for rice producers, distributors, and traders, offering improved quality control, cost optimization, and supply chain management. They also hold the potential to aid in the development of rice varieties that are more resistant to disease, pests, and environmental stress. While most existing studies in the rice classification domain rely on traditional machine-learning techniques that necessitate feature extraction engineering processes, our research explores the effectiveness of novel deep-learning models for this task. We evaluated the performance of various contemporary deep-learning models, including Residual Network (ResNet), Visual Geometry Group (VGG) network, EfficientNet, and MobileNet. These models were tested on a dataset comprising 75,000 images, classified into five different rice categories. We assessed each model using established evaluation metrics such as accuracy, F1 score, precision, recall, and per-class accuracy. Our findings showed that the EfficientNet-based model delivered the highest accuracy (99.67%), while the MobileNet-based model excelled in the speed of classification (2556 s). We concluded that, compared to traditional machine learning methods, the models employed in our study are highly scalable and capable of managing large volumes of complex data with millions of features and samples.http://www.sciencedirect.com/science/article/pii/S2666154323003976Rice grain classificationDeep learningMachine learningAutomatic feature extraction |
| spellingShingle | Farshad Farahnakian Javad Sheikh Fahimeh Farahnakian Jukka Heikkonen A comparative study of state-of-the-art deep learning architectures for rice grain classification Journal of Agriculture and Food Research Rice grain classification Deep learning Machine learning Automatic feature extraction |
| title | A comparative study of state-of-the-art deep learning architectures for rice grain classification |
| title_full | A comparative study of state-of-the-art deep learning architectures for rice grain classification |
| title_fullStr | A comparative study of state-of-the-art deep learning architectures for rice grain classification |
| title_full_unstemmed | A comparative study of state-of-the-art deep learning architectures for rice grain classification |
| title_short | A comparative study of state-of-the-art deep learning architectures for rice grain classification |
| title_sort | comparative study of state of the art deep learning architectures for rice grain classification |
| topic | Rice grain classification Deep learning Machine learning Automatic feature extraction |
| url | http://www.sciencedirect.com/science/article/pii/S2666154323003976 |
| work_keys_str_mv | AT farshadfarahnakian acomparativestudyofstateoftheartdeeplearningarchitecturesforricegrainclassification AT javadsheikh acomparativestudyofstateoftheartdeeplearningarchitecturesforricegrainclassification AT fahimehfarahnakian acomparativestudyofstateoftheartdeeplearningarchitecturesforricegrainclassification AT jukkaheikkonen acomparativestudyofstateoftheartdeeplearningarchitecturesforricegrainclassification AT farshadfarahnakian comparativestudyofstateoftheartdeeplearningarchitecturesforricegrainclassification AT javadsheikh comparativestudyofstateoftheartdeeplearningarchitecturesforricegrainclassification AT fahimehfarahnakian comparativestudyofstateoftheartdeeplearningarchitecturesforricegrainclassification AT jukkaheikkonen comparativestudyofstateoftheartdeeplearningarchitecturesforricegrainclassification |